the nonprofit board as a team - adea.com.au · general corporate governance literature recognizes...

NONPROFIT MANAGEMENT & LEADERSHIP, vol. 22, no. 4, Summer 2012 © 2012 Wiley Periodicals, Inc. 461Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/nml.21040

RESEARCH NOTES

The Nonprofit Board as a Team

Pilot Results and Initial Insights

Gavin Nicholson, Cameron Newton,Myles McGregor-Lowndes

Queensland University of Technology

There is a growing desire for boards of nonprofits to deliver bet-ter governance to the organizations they control. Consequently,self-evaluation has become an important tool for nonprofitboards to meet these expectations and demonstrate that they aredischarging their responsibilities effectively. This articledescribes initial results aimed at developing a psychometricallysound, survey-based board evaluation instrument based on theTeam Development Survey, which assesses the team attributesof an organization’s board. Our results indicate that while con-structs applicable to teams generally appear to apply to boards,there are also important differences. We highlight how a per-ception of board objective clarity, appropriate skills mix,resource availability, and psychological safety were positivelyand significantly associated with measures of board, manage-ment, and organizational performance.

Keywords: board evaluation, boards of directors, nonprofit governance

AN ORGANIZATION’S BOARD CAN have profound effects on bothorganizational outcomes and personal well-being. At oneend of the spectrum, there is an increasing recognition of

Correspondence to: Gavin Nicholson, Queensland University of Technology, Aus-tralian Centre of Philanthropy and Nonprofit Studies, GPO Box 2434, Brisbane,Queensland 4001, Australia. E-mail: [email protected]

the positive roles that a board can play in creating value for the or-ganization it governs (for example, Huse, 2007), particularly innonprofit boards (for example, Brown, 2002, 2005; Herman andRenz, 1998; Jackson and Holland, 1998). At the other end are nu-merous well-documented instances of board failure affecting orga-nizational performance (see Carman, 2011) as well as a string ofsignificant cases highlighting potentially disastrous results for theindividuals involved (see McGregor-Lowndes, 1995, discussing the National Safety Council case in Australia). The relationship be-tween boards and organizational performance does not appear re-lated to any single context; for instance, board failure transcendsnational boundaries and regulatory systems (consider the cases ofthe National Safety Council in Australia; the Singapore KidneyFoundation in Singapore; the Foundation for New Era Philanthropyin the United States; and Moonbeams in the United Kingdom).

Consequently there is an increasing interest in nonprofitcapacity building, both in general and for the board of directorsin particular (Nobbie and Brudney, 2003). Given the broad con-clusions from a global study that ineffective nonprofit boards are linked to weak organizational accountability (Salamon andChinnock, 2004), it becomes apparent that nonprofit boards willincreasingly be called to ensure they are performing at their best.If not, it is likely that mandated public reporting practices will fol-low, along with the added complexity, red tape, and cost (Smithand Richmond, 2007).

An important challenge facing boards seeking to developcapacity is appropriate diagnosis of their weaknesses. Despite theincreasing awareness of the importance of an effective board and aplethora of normative advice, surprisingly little has been done toensure the reliable diagnosis of problems facing nonprofit boards,particularly outside of North America. Mirroring research on theirfor-profit counterparts (for example, Finkelstein and Mooney,2003), research conducted into nonprofit boards reveals a generalbelief that significant improvements are possible (Morrow and Bartlett, 2007) but that measurement and diagnosis across abroad swath of organizations is difficult (for example, Nicholsonand Newton, 2010).

In this article we present initial results relating to the develop-ment of a new diagnostic tool for nonprofit boards and show thatfocusing on the nonprofit board as a team appears to be a fruitfulavenue for continued research. First, we present the logic behind ourapproach, including a brief overview of the growing interest in thebehavioral aspects of corporate governance and how the board oper-ates as a group or team. We then briefly highlight peer-reviewed diag-nostic tools currently available to the sector before presenting ourresults and concluding with the implications of our findings forresearch and practice.

462 NI C H O L S O N, NE W T O N, MCGR E G O R-LO W N D E S

Nonprofit Management & Leadership DOI: 10.1002/nml

Despite theincreasing

awareness of theimportance of aneffective board

and a plethora ofnormative advice,surprisingly littlehas been done to

enable thereliable diagnosis

of problemsfacing nonprofit

boards,particularly

outside NorthAmerica.

Developing a Team-Focused Diagnostic for Nonprofit Boards

A challenge for any governance diagnostic tool in the nonprofitarena is the diversity and range of contexts (for example, size, cul-ture, industry, and so on) in which it will be used. Nonprofits varyfrom flat, feminist-inspired structures to church theocratic bureau-cracies that take many different legal forms. This means differentnonprofits will have different legal requirements and expectationsas to their tasks, responsibilities, and attitudes to governance. Forexample, a recent survey of a narrow sample of health and alliednonprofits in one region in Australia identified six different legalstructures in use (Nicholson, Newton, and Sheldrake, 2008;see Figure 1). Consequently, the requirements and even operating

TH E NO N P R O F I T BO A R D A S A TE A M 463


Other1%

Incorporated Association84%

Trust0.4%Royal Charter or

Letters Patent1%

Company limitedby guarantee

3%

UnincorporatedAssociation

4%

Unsure6%

Legislation1%

Figure 1. Legal Forms of Nonprofit Health Organizations in One Australian State

procedures for governing bodies is likely to be highly variable ifviewed primarily from a legal perspective.

Growing interest in group effectiveness at a board level (Finkelsteinand Mooney, 2003; Forbes and Milliken, 1999; Hillman and Dalziel, 2003; Nicholson and Kiel, 2004) and human behaviorwithin boards and governing groups (for example, Leblanc, 2005;Pye, 2005; van Ees, Gabrielsson, and Huse, 2009) offers an approachto address this challenge. This strong and growing tradition in thegeneral corporate governance literature recognizes that good gover-nance relies on the behavior of people as much as the law (for exam-ple, Sonnenfeld, 2002) and that the phenomenon of effectiveorganizational governance is situated in multiple levels: It is an orga-nizational, group, and individual phenomenon (Dalton and Dalton,2011).

The multidisciplinary recognition of the group-based nature ofeffective corporate governance is another important advantage offocusing on group performance in the articulation and measurementof effective governance. For instance, legal scholars are clear that it isthe board (or group), not individual directors, with the power tomake decisions (Bainbridge, 2003), while other major disciplines,including sociology (Pfeffer, 1973), psychology (Davis, Schoorman,and Donaldson, 1997), and economics (Eisenhardt, 1989), highlightimportant behavioral factors thought to influence good board deci-sion making and effective personal action. Quite simply, there isstrong and growing evidence that the effectiveness of an organiza-tion’s governance system relies on an effective board operating welltogether as a team. Thus, many view opening the “black box” of howthe board operates as the most important challenge facing the field(Daily, Dalton, and Cannella, 2003; Huse, 2005).

By focusing on the board as a group or team, we are clearlydelimiting organizational performance from governance task (orboard role) performance. Board role and organizational performancedo not form part of our primary research focus in this article due toimportant empirical and theoretical considerations. We do notinclude organizational performance for two important reasons. First,there is a long mediation process between what boards do and effec-tive organizational performance—things like the business environ-ment, management, and luck matter (Hillman and Dalziel, 2003;Nicholson and Kiel, 2004). Second, the problems associated withcomparing performance across a broad range of nonprofit organiza-tions are well recognized (Herman and Renz, 1997, 2004) and arebeyond the scope of a psychometrically valid self-diagnosis tool forboards.

Similarly, board role execution (or task performance) is a par-ticularly difficult subject to diagnose and measure. Again, we excludeit from our diagnostic tool for two important reasons. First, defini-tional problems abound, with different authors employing differenttypologies (for example, compare Hillman and Dalziel, 2003; Hung,



There is strongand growing

evidence that theeffectiveness of an

organization’sgovernance

system relies onan effective board

operating welltogether as a

team.

1998; and Johnson, Daily, and Ellstrand, 1996), and emerging evi-dence suggests that these typologies do not match how governorsand managers think about their roles (Nicholson and Newton,2010). A major insight of the general groups literature is that thegroup finds it difficult to assess its team product—that is, what the team does. The alternative is to seek external measures of boardrole performance, a difficult and complex task that to date has pro-vided different assessments of performance from different stake-holders (Herman and Renz, 1997, 2004). In summary, there aresignificant conceptual and practical difficulties in measuring boardrole and organizational performance.

In contrast, there is significant evidence that groups can validlymeasure their internal attributes (Hackman, 2002). Our approach buildson the group effectiveness research agenda of Hackman, Wageman,and colleagues (Hackman, 2002; Hackman and Wageman, 2005a,2005b). Specifically, we utilize concepts and elements of their teamdevelopment model and its associated instrument, the Team Devel-opment Survey (TDS) (Wageman, Hackman, and Lehmann, 2005).The TDS model provides a strong empirical and theoretical basis forunderstanding the social forces at work in boards, and we haveadapted its key elements to assess group-based attributes of nonprofitboard governance in the Australian context. Our approach involveddeveloping new constructs and items, revising existing items, andremoving constructs not relevant to our context. A full list of theitems (grouped by construct) is provided in the Appendix to this arti-cle, where an italicized item indicates a direct derivation from Wageman, Hackman, and Lehmann (2005).

Comparison with Existing Instruments for Nonprofit Governance Diagnosis

Ensuring that a governing body or team functions appropriatelyand effectively necessitates evaluation. Yet there are few validatedtools to help governing bodies evaluate their work. Most relevant tononprofits are the Slesinger self-assessment tool (Slesinger, 1991),the Board Self-Assessment Questionnaire (BSAQ) (Holland, 1991), and the Governance Self-Assessment Checklist (GSAC)(Gill, Flynn, and Reissing, 2005), North American tools that havebeen subject to significant peer review.

These tools have made valuable contributions to our under-standing of nonprofit governance; however, there are three impor-tant reasons why an open, validated, group-focused diagnostic forboards would build on this early work. First, the current tools donot provide a clear separation between organizational performanceand board effectiveness. For instance, Slesinger’s (1991) board evalua-tion tool, explored by Herman and Renz (1997, 1998, 2004), provideshigh correlations between board and organizational performance,



suggestive of a mixing of the two concepts. This is corroborated bythe general conclusion of Herman and Renz (1998) that “both boardmembers and chief executives apparently regard the financial con-dition of the organization as the true measure of board effectiveness”(p. 700). Although this may be the perception of their performanceby those involved in governance (put another way: their perfor-mance is organizational performance, and generally the singledimension of financial performance), this does not mean that it isaccurate. For example, it might not be what others or society wouldperceive. The different ratings of performance provided by differentstakeholders support this conclusion. Given the overlap betweenperceptions of organizational and board performance reported here,there are significant conceptual difficulties with unidimensional (orsingle-factor) measures of board performance (as reported in theother diagnostic instruments we review later). As the academic lit-erature clearly separates board and organizational effectiveness (forexample, Dalton and Dalton, 2011; Huse, 2007), addressing thisissue is an important concern for boards seeking to understand theirperformance.

A second issue common to all the board evaluation tools is thelack of clear empirical support for the theoretical structure of themeasures of board effectiveness they posit. Slesinger’s self-assessmenttool (1991), the BSAQ (Holland, 1991), and the GSAC (Gill, Flynn,and Reissing, 2005) each report that all items load on the single factor of board effectiveness. Yet the diagnostics are positioned asmeasuring different dimensions of board effectiveness. For instance,Slesinger’s tool has been adapted to an eleven-dimension, single-construct model (Herman and Renz, 1997, 1998, 2004); the BSAQinvolves six dimensions of an effective board (Holland, 1991); andthe GSAC (Gill, Flynn, and Reissing, 2005) has twelve dimensionsbut is explained only as a single construct of board effectiveness.

This point is critical in a diagnostic tool, as multidimensional-ity suggests, either explicitly or implicitly, that boards need actionto improve their performance on the dimensions provided to effectoverall performance changes. The minimal evidence that thesedimensions are validly measurable (and the lack of exploration ofthe relationships between the various dimensions) suggests that fur-ther work could improve our understanding of boards and provisionof interventions. For instance, the dimensionality of Slesinger’s(1991) tool does not appear to have been fully investigated. TheBSAQ appears to mix the conceptually different categories of preciseboard role performance (for example, strategic skills, political skills)with group (for example, interpersonal) and individual-based mea-surements (for example, analytic, education). Similarly, the GSAC,while reporting differences in means across the dimensions, provides144 items in twelve scales (average twelve items per scale) reflectedin very high reliability scores (Cronbach’s alpha), which suggeststhat items could easily be dropped to improve the tool. Put simply,



psychometric testing suggests that the more items included in a sur-vey, the less clear are the underlying constructs.

When current tools are taken together, it seems that they maynot have the discriminant validity necessary to inform boards onwhat aspect of their performance they should improve. This is bestdone by reporting correlation tables and other results that highlightthe relationships between variables (for example, factor analyses) asimportant steps required to establish the structure of board effec-tiveness and its valid measurement. Without these steps, it is possi-ble that boards are being directed to assess the wrong aspects of theirperformance or (more likely) that they measure some dimensions ofeffectiveness more validly than others. To advance the field in a con-sistent and coherent fashion, end users and researchers need tounderstand these aspects of any diagnostic instrument.

A third challenge is one common to academic work, particularlywith respect to boards: limitations arising from technical issues.Good psychometric measurement requires one of two approaches.In one approach, the researcher can concentrate on a very narrowaspect of measurement, such as a single subdimension of effective-ness. In the current context, this would allow for a deep but narrowunderstanding of an aspect of board effectiveness. An alternativeapproach is to undertake a committed program of research where thetools develop as researchers gain a better understanding of the con-cepts being studied and the items being used to measure these concepts across multiple concepts. Current diagnostics typically favorthe second approach and could be improved through greater consis-tency in operationalization and application to more general populations.Perhaps the most rigorous development documentation is providedby the BSAQ, where the number of items varied across the devel-opment cycle. The original article reports sixty-nine items (Holland, 1991); the second, seventy-three items (Jackson and Holland,1998); the third did not report the number of items (Holland andJackson, 1998); and the fourth reported thirty-seven items (Brown,2005). Given the range of reliability scores reported across the stud-ies, the lack of detail on what or why items were removed, and theemphasis on a single dimension of effectiveness in the results, thereare important limitations to any claims made about how the tool canguide board improvement. Similarly, several samples appear to bequite atypical and would benefit from a broader sample frame if theresults are to be applied in a wider governance setting.

Overall, while the peer-reviewed instruments available to boardsand researchers are a positive step, they lack the discrimination nec-essary to highlight best the aspect of performance that a board needsto improve or address. These diagnostics all tend to focus on “over-all” board performance and have not reliably established the requi-site dimensionality that could guide board change. To use ametaphor, while it is important for a doctor to tell patients they aresick, it is more useful to tell them what part of their physiology is



affected and how one symptom relates to another. The focus on over-all board performance also introduces the problem that the diag-nostics presented here, to varying degrees, fail to differentiatebetween organizational and board performance. Given expected timelags between board action and performance and possible interactionswith multiple factors (including management performance), this isproblematic. For instance, if management is performing well, orga-nizational performance measures may not reveal that a board is per-forming poorly; it is only when management changes that theseboard deficiencies will flow through to organizational performance(Nicholson and Kiel, 2004). Thus, measures that conflate board andorganizational performance could be misleading. Finally, there aresome unique samples used in the research that would appear to dif-fer substantially from many nonprofit contexts.

Conceptualizing the Board as a TeamOne way to move beyond a unitary understanding of board effec-tiveness is to recognize the importance of team structure toboard—and therefore governance—performance (Bainbridge,2002; Forbes and Milliken, 1999; Huse, 2007). Focusing on theboard as a team also addresses problematic measurement and va-lidity concerns posed by differing legal structures and compliancerequirements of directors both within and between countries.

Since a board has important differences from traditional workgroups in terms of structure and power, there are strong theoreticaland practical reasons for developing a tool specifically designed forthis context. First, unlike other work groups, boards are sui generis(or a separate legal entity) (Bainbridge, 2002) with members notreporting to a superior but elected by members and with equal for-mal power. Second, boards tend to be larger, composed of more out-siders, and meet more episodically than traditional work groups(Forbes and Milliken, 1999). These differences tend to be even morepronounced in the nonprofit sector.

Our basic model follows the essential elements of Wageman,Hackman, and Lehmann’s (2005) TDS model, adapted for the uniquenature of the governing body. It consists of five components, illus-trated in Figure 2. As the figure indicates, we do not propose spe-cific relationships between the model components or that thesecomponents, in themselves, ensure effectiveness but rather believethat they provide the foundation for effective execution of gover-nance tasks.

First, we examine whether there is a clear delineation in theboundaries of the governance group, a widely cited attribute of aneffective board (for example, Carter and Lorsch, 2004; Fishel, 2008;Kiel and Nicholson, 2003). This aspect of the instrument aims to mea-sure whether board members (1) know who is on their “governance



team”; (2) know what they need to do; and (3) work together as agroup. This component of the instrument is based on Wageman,Hackman, and Lehmann’s (2005) concept of a real team and an ele-ment of their model that deals with compelling direction.

Second, we address the widely acknowledged challenge of ensur-ing an appropriate mix of talent around the board table (for exam-ple, Hough, McGregor-Lowndes, and Ryan, 2006; Nicholson,Newton, and Sheldrake, 2008). This component of the instrumentrecognizes that there are many ways to organize the governing body;for that reason, it does not prescribe a “best” standard for governingboard composition. For instance, it does not assess boards againstany ideal list or mix of skills. Instead, the instrument seeks toprompt boards to reflect on how appropriate they see the board’scomposition in three different dimensions. Items covering the levelof diversity and experience base require board members to assesscomposition in light of their specific context.

Third, we aim to identify whether the board has a clear visionfor its role and the organization. Defining what is strategy and a gov-erning body’s role in the strategy development process is beyond thescope of this article (see, for example, Hendry, Kiel, and Nicholson,2010; Stiles and Taylor, 2001, for reviews of the board’s role in strat-egy). Instead, we follow Wageman, Hackman, and Lehmann (2005)to assess whether the board finds its work sufficiently challengingand engaging. If the board’s work is not challenging and engaging,the board is likely to lack a strategic focus and not engage all boardmembers. The instrument seeks to measure whether the boardbelieves it is in control of the organization and pursuing a worth-while direction.

The fourth aspect of our model examines whether the board hasthe requisite information and processes to operate effectively. Thiselement of the survey concentrates on whether a board’s processesand resources are adequate and appropriate. It seeks to measure the



If the board’swork is not

challenging andengaging, the

board is likely tolack a strategicfocus and not

engage all boardmembers.

Supporting processes

Defined team

Talent to task

Agreed direction

Learning and growth

Board roleperformance

Organizationalperformance

Figure 2. Board as a Team: How the Board’s Team Structure Contributes to Organizational Performance

views of board members about the adequacy and timeliness of theinformation they receive and if the governing body may require anyother material resources to carry out its role.

Finally, an effective group needs to continue to evolve, and sothe final element of the survey involves assessing the feedback mech-anisms in the board. We particularly focus on peer-to-peer support,aspects of teamwork, group norms, and the nature of the relation-ships between board members and between the board and the man-agement team.

Diagnostic Instrument and Early ResultsThe aim of this project is to develop a psychometrically sound in-strument to assist nonprofit boards in measuring the team-basedaspects of their performance. In this article we present initial re-sults for the team-based constructs that were adopted and adaptedto ensure contextual relevance to nonprofit boards. To improveface and construct validity, we circulated the draft instrument to apanel of governance practitioners. Unless we had strong theoreti-cal reasons to the contrary, we made further changes and refine-ments to the instrument based on their responses prior toconducting the survey.

Participants and ProcedureIn total, 118 active nonprofit board members from eighteen boardsaround Australia took part in the survey. Participants served onboards drawn from industries including education, research,health, and social services. The mean size of each board was 9.9members (standard deviation [SD] � 4.82). Board members wereinvited via invitation or word of mouth to participate in the survey.The chair of each board was provided with a code so that we couldidentify the board to which each participant belonged. This codewas provided to each board member, who was then able to accessthe relevant survey via a secure Internet survey system. As part ofthis process, each board member also generated a unique codeknown only to him or her, allowing future tracking of responsesacross different survey instruments and over time.

Preliminary ResultsAs the survey is still in its pilot phase, we are unable to provide ex-haustive modeling of the data to explore in-depth multivariate re-lationships. However, we can present preliminary data analysesrelated to each construct and also consider the relationships ofthese constructs to subjectively rated performance outcomes to ex-amine the relative efficacy of the team-based focus of the instru-ment. Table 1 presents means, SDs, and intercorrelations for the



Tab

le 1

. Des

crip

tive

Sta

tist

ics

and

Cro

nba

ch’s

Alp

ha

Coe

ffic

ien

ts o

f F

ocal

Var

iabl

es

Var

iabl

eM

(SD

)1

23

45

67

89

1011

1213

14

1. D

efin

ed3.

43 (

0.95

)(0

.74)

2. I

nte

rdep

ende

nt

3.54

(0.

75)

0.18

(0.6

1)3.

Cle

ar3.

86 (

0.88

)0.

19*

0.38

**(0

.75)

4. C

onse

quen

tial

4.14

(0.

74)

0.20

*0.

42**

0.44

**(0

.73)

5. D

iver

sity

3.78

(0.

72)

0.27

**0.

35**

0.54

**0.

30**

(0.7

2)6.

Ski

lls

3.78

(0.

78)

0.24

**0.

27**

0.59

**0.

31**

0.71

**(0

.75)

7. I

nfo

rmat

ion

3.85

(0.

55)

0.08

0.06

0.50

**0.

41**

0.21

*0.

35**

(0.7

1)8.

Res

ourc

es3.

85 (

0.65

)0.

14�

0.00

0.51

**0.

180.

27**

0.34

**0.

66**

(0.6

5)9.

Tas

k 3.

78 (

0.73

)0.

22*

0.50

**0.

56**

0.38

**0.

55**

0.68

**0.

37**

0.37

**(0

.90)

10. U

nh

elpf

ul

2.21

(0.

65)

�0.

130.

10�

0.12

�0.

000.

04�

0.10

�0.

21*

�0.

050.

07(0

.84)

11. P

sych

olog

ical

4.

05 (

0.56

)0.

21**

0.38

**0.

57**

0.42

**0.

48**

0.54

**0.

46**

0.40

**0.

70**

�0.

09(0

.83)

safe

ty12

. Org

aniz

atio

n

4.03

(0.

79)

0.20

**0.

20**

0.58

**0.

27**

0.42

**0.

41**

0.37

**0.

47**

0.46

**0.

080.

46**

perf

orm

ance

13. B

oard

3.

84 (

0.86

)0.

26**

0.29

**0.

69**

0.36

**0.

53**

0.67

**0.

46**

0.51

**0.

64**

�0.

050.

60**

0.71

**

perf

orm

ance

14. M

anag

emen

t 4.

29 (

0.62

)0.

090.

090.

48**

0.10

0.32

**0.

31**

0.27

**0.

41**

0.19

*�

0.13

0.28

**0.

63**

0.56

**

team

per

form

ance

15. B

oard

–man

agem

ent

4.15

(0.

80)

0.20

*0.

30**

0.56

**0.

46**

0.36

**0.

45**

0.49

**0.

46**

0.51

**�

0.00

0.56

**0.

60**

0.65

**0.

50**

coll

abor

atio

n

*p�

0.05

.

**p

�0.

01.

constructs related to the board as a team, psychological safety, andthe single-item organizational performance measures. The con-structs and example items are presented in the Appendix to thisarticle.

The diagonal of Table 1 displays the reliability coefficients forthe multi-item constructs. Inspection of the table reveals that themajority of scales satisfied the generally accepted threshold for inter-nal reliability using Cronbach’s (1951) alpha coefficient (i.e., alpha �0.70). Two other variables, “resources” and “independent,” wereabove 0.60 and were retained, given the exploratory nature of thisresearch. Three other original variables (“challenge,” “stable,” and“set by the board”) had alpha coefficients that fell well below thethreshold and were excluded from further analysis. For these vari-ables, we have developed additional items or reworded existing itemsfor further analysis in future surveys.

Overall, the results are generally favorable for this part of thesurvey development process. The significant correlations betweenvariables are logically interpretable and generally in line with expec-tations. In addition, those variables that might be considered inde-pendent variables in future models are not correlated so highly as topresent potential multicollinearity issues in any future research.

Tests of the Predictive Ability of the ModelFour multiple regression analyses were performed to investigatethe predictive ability of the team-based model and psychologicalsafety on self-reported performance. Performance variables as-sessed board, organizational, board–management, and manage-ment team performance. All team variables and psychologicalsafety were entered on step 1 simultaneously. The results are dis-played in Table 2.

As reported in Table 2, entry of the team variables and psycho-logical safety at step 1 accounted for significant increments in vari-ance for ratings of board (R2 � 0.64, F(11,102) � 16.59, p � 0.001),management team (R2 � 0.31, F(11,98) � 4.01, p � 0.001),board–management collaboration (R2 � 0.48, F(11,100) � 8.22,p � 0.001), and organizational (R2 � 0.43, F(11,102) � 7.09, p �0.001) performance. The value of the model we propose is supportedby the differences in significance between constructs seen acrossthese analyses. For instance, board performance was significantlyand logically predicted by clarity of board objectives (b� 0.30, p �0.01), appropriate skills of board members (b� 0.27, p � 0.05), andboard resource availability (b � 0.18, p � 0.05). For managementteam performance, clarity (b � 0.34, p � 0.05) and resources (b �0.29, p � 0.05) were most important, and for board–managementcollaboration performance, team psychological safety was the mostsignificant driver (b� 0.23, p � 0.05). Last, from an organizationalperformance perspective, the key team-level predictor was the clear



objective of the board (b � 0.39, p � 0.01). Overall, these resultspoint to differential importance and prediction of key team-level andpsychological variables that are vital to consider in assessing and advis-ing on the many facets of board and organizational performance.

DiscussionThis article seeks to respond to the increasing emphasis placed onboard effectiveness in the nonprofit sector by outlining initial re-sults for a diagnostic tool designed to assess the team-based as-pects of a board’s performance. We chose a team focus for thediagnostic tool for three main reasons: (1) it is likely applicableacross many different legal and contextual differences facing non-profit boards; (2) it answers many calls in the general governanceliterature to better understand the nature of the board’s work (forexample, Daily, Dalton, and Cannella, 2003; van Ees, Gabrielsson,and Huse, 2009); and (3) it provides feedback to boards on variousaspects of their activities that are clearly differentiated from organi-zational performance and within their ability to change.

Our first major conclusion is that conceptualizing boards as ateam appears to hold a great deal of promise. Although we neededto adapt and expand the well-recognized TDS diagnostic survey(Wageman, Hackman, and Lehmann, 2005), the various constructsfrom the team-based diagnostic survey performed well in the firstmajor test. The reliability statistics reported here, together with thestrong explanatory power for ratings of board and organizational



Table 2. Multiple Regression: Board as a Team Characteristics and Performance Outcomes

Independent Board Management Team Board–Management OrganizationalVariables Performance Performance Collaboration Performance Performance

Defined 0.07 �0.04 0.07 0.07Interdependent �0.05 0.05 0.03 �0.11Clear 0.30*** 0.34** 0.19 0.39***

Consequential 0.03 �0.10 0.13 0.02Diversity �0.02 0.17 �0.09 0.10Skills 0.27** 0.08 0.09 �0.07Information �0.03 �0.08 0.15 �0.05Resources 0.18** 0.29** 0.11 0.20*

Task 0.16 �0.30* 0.06 0.12Unhelpful 0.03 �0.06 0.11 0.14Psychological safety 0.12 0.12 0.23** 0.12*

R2 0.64*** 0.31*** 0.48*** 0.43***

*p � 0.10.

**p � 0.05.

***p � 0.01.

effectiveness, suggest that the tool is a reliable measure of things thatmatter to board effectiveness. At the same time, the differencesreported (in both the correlation results and subsequent regressionanalyses) suggest that the constructs we measure are significantlydifferent and appear to influence different aspects of performance indifferent ways. Initial results also suggest that some aspects of theteam-based approach (notably clarity in objectives) appear to beglobal in their effect on perceptions of performance while other con-structs may point to different elements of governance effectiveness(for example, how well the board works with management or evenstakeholders).

A second major conclusion is that while conceptualizing theboard as a team may be a useful frame, our initial findings also indi-cate that wholesale application of the general small-groups literatureto boards may not be appropriate. Our findings suggest that someaspects of how boards operate may differ from general team models,possibly due to the important differences between a board and othersmall groups typically examined in the business context (Forbes andMilliken, 1999). For instance, the low reliability of the interdepen-dent scale suggests that this concept may not be so applicable toboards. This makes sense: The episodic nature of the board’s workvaries dramatically from most work groups that meet on a far moreregular basis. Similarly, the heavy emphasis on a singular group inter-action in the formal environment of a board meeting may well meanthat we need to rethink how board members work together ratherthan just apply the findings from the groups literature.

A third major conclusion from the research is that the aspects ofgroup performance we measured appear to be related with percep-tions of performance, at both the organizational and the board level.This is important if we are to understand better how the board’swork contributes to overall organizational outcomes.

Implications for PracticeThe insights from this research will, we hope, prove useful forboards, their advisers, and their regulators. First, the applicability ofteam-based diagnostic tools to the board suggests that group-basedinterventions may prove useful in developing an effective board. Forinstance, the strong relationship between clarity in board objectivesand various aspects of perceived importance corroborates regula-tory guidelines and practitioner advice to ensure there is a clearsense of agreed purpose for the board and its role.

Second, the diagnostic tool that we have developed provides anadditional resource for those boards seeking to improve their per-formance. Specifically, it isolates a reliable way to measure aspects ofwhat they do, so that the group can concentrate energies more appro-priately on the aspects of performance that require development.



Although further research will, we hope, provide more guidance onwhich aspects of team performance are associated with which aspectsof board and organizational performance, this is nevertheless animportant step.

Limitations and Future ResearchThis study provides a promising start, but there are clear limita-tions to our results. First, there is a limited sample size (n � 118)in a specific context (Australian nonprofits). In addition, the sam-ple was drawn from organizations that volunteered to participate,which may have introduced sample bias into the study. Future re-search could concentrate on broadening and deepening the sampleso as to ensure the results are generalizable.

A second key limitation is that we have only demonstrated therelationship between our diagnostic tool and perceptions of perfor-mance. Although this is an important step, understanding the rela-tionship between the constructs in the model and more objectivemeasures of performance (for example, stakeholder perceptions,financial performance) would provide more clarity around the use-fulness of the tool. It would also overcome possible problems ofcommon method variance that might have influenced our results.

Despite these limitations, our findings provide a clear path fora group-based or behavioral approach to studying boards of direc-tors. Many other aspects of group performance that we have notincluded in this study are worthy of investigation (such as conflictresolution behaviors). Similarly, the multilevel nature of boards (thatis, board members and managers are individuals who come togetheras a single group) poses exciting challenges and opportunities forthe study of boards. Finally, identifying the differences and similar-ities between different board contexts and aspects of group perfor-mance appears to hold promise for further research.

ConclusionAll groups require feedback if they are to improve their perfor-mance (Sonnenfeld, 2002). A key challenge for nonprofit boardshas been sourcing a rigorous, appropriate way of gaining feed-back about issues that they can influence directly. The diagnostictool outlined in this article provides an empirically based frame,applicable across the vast majority of nonprofit boards. It offersthe first step in an alternative to tools sourced from Australianfor-profit products or from a different culture and is the subjectof a careful and thorough approach to psychometric validation.We hope it will lead to valid and insightful feedback for non-profit boards and, ultimately, better outcomes for them and theirorganizations.



A key challengefor nonprofit

boards has beensourcing arigorous,

appropriate wayof gaining

feedback aboutissues that theycan influence

directly.

References

Bainbridge, S. M. “Why a Board? Group Decision Making in Corpo-rate Governance.” Vanderbilt Law Review, 2002, 55, 1–55.

Bainbridge, S. M. “Director Primacy: The Means and Ends of Cor-porate Governance.” Northwestern University Law Review, 2003,97, 547–606.

Brown, W. A. “Composition and Performance of Nonprofit Boardsof Directors.” Journal of Applied Management and Entrepreneurship,2002, 7, 43–57.

Brown, W. A. “Exploring the Association between Board and Orga-nizational Performance in Nonprofit Organizations.” NonprofitManagement and Leadership, 2005, 15(3), 317–339.

Carman, J. G. “What You Don’t Know Can Hurt Your Community:Lessons from a Local United Way.” Nonprofit Management andLeadership, 2011, 21(4), 433–447.

Carter, C. B., and Lorsch, J. W. Back to the Drawing Board. Boston:Harvard Business School Press, 2004.

Cronbach, L. J. “Coefficient Alpha and the Internal Structure ofTests.” Psychometrika, 1951, 16, 297–335.

Daily, C. M., Dalton, D. R., and Cannella, A. A. “Corporate Gover-nance: Decades of Dialogue and Data.” Academy of ManagementReview, 2003, 28(3), 371–382.

Dalton, D. R., and Dalton, C. M. “Integration of Micro and MacroStudies in Governance Research: CEO Duality, Board Composi-tion, and Financial Performance.” Journal of Management, 2011,37(2), 404–411.

Davis, J. H., Schoorman, F. D., and Donaldson, L. “Toward a Stew-ardship Theory of Management.” Academy of Management Review,1997, 22(1), 20–47.

Eisenhardt, K. M. “Agency Theory: An Assessment and Review.”Academy of Management Review, 1989, 14, 57–74.

Finkelstein, S., and Mooney, A. C. “Not the Usual Suspects: How toUse Board Process to Make Boards Better.” Academy of ManagementExecutive, 2003, 17, 101–113.

Fishel, D. The Book of the Board: Effective Governance for Non-profitOrganisations. Annandale, New South Wales: Federation Press,2008.

Forbes, D. P., and Milliken, F. J. “Cognition and Corporate Gover-nance: Understanding Boards of Directors as Strategic Decision-Making Groups.” Academy of Management Review, 1999, 24(3),489–505.

Gill, M., Flynn, R. J., and Reissing, E. “The Governance Self-AssessmentChecklist: An Instrument for Assessing Board Effectiveness.” Non-profit Management and Leadership, 2005, 15, 271–294.

Hackman, J. R. Leading Teams: Setting the Stage for Great Perfor-mances. Boston: Harvard Business School Press, 2002.



Hackman, J. R., and Wageman, R. “A Theory of Team Coaching.”Academy of Management Review, 2005a, 30(2), 269–287.

Hackman, J. R., and Wageman, R. “When and How Team LeadersMatter.” Research in Organizational Behavior, 2005b, 26, 37–74.

Hendry, K. P., Kiel, G. C., and Nicholson, G. “How Boards Strategise:A Strategy as Practice View.” Long Range Planning, 2010, 43,33–56.

Herman, R. D., and Renz, D. O. “Multiple Constituencies and theSocial Construction of Nonprofit Organization Effectiveness.” Non-profit and Voluntary Sector Quarterly, 1997, 26, 185–206.

Herman, R. D., and Renz, D. O. “Nonprofit Organizational Effec-tiveness: Contrasts between Especially Effective and Less EffectiveOrganizations.” Nonprofit Management and Leadership, 1998, 9(1),23–38.

Herman, R. D., and Renz, D. O. “Doing Things Right: Effectivenessin Local Nonprofit Organizations, a Panel Study.” Public Adminis-tration Review, 2004, 64(6), 694–704.

Hillman A. J., and Dalziel, T. “Boards of Directors and Firm Perfor-mance: Integrating Agency and Resource Dependence Perspec-tives.” Academy of Management Review, 2003, 28(3), 383–396.

Holland, T. P. “Self-Assessment by Nonprofit Boards.” Nonprofit Man-agement and Leadership, 1991, 2(1), 25–36.

Holland, T. P., and Jackson, D. K. “Strengthening Board Performance:Findings and Lessons from Demonstration Projects.” NonprofitManagement and Leadership, 1998, 9(2), 121–134.

Hough, A. D., McGregor-Lowndes, M., and Ryan, C. M. “The Train-ing Grounds of Democracy? Social Trends and Nonprofit Gover-nance.” Working Paper No. CPNS 31. Brisbane: Australian Centrefor Philanthropy and Nonprofit Studies, 2006.

Hung, H. “A Typology of the Theories of the Roles of GoverningBoards.” Corporate Governance, 1998, 6(2), 101–111.

Huse, M. “Accountability and Creating Accountability: A Frameworkfor Exploring Behavioural Perspectives of Corporate Governance.”British Journal of Management, 2005, 16, S65–S79.

Huse, M. Boards, Governance and Value Creation. London: CambridgeUniversity Press, 2007.

Jackson, D. K., and Holland, T. P. “Measuring the Effectiveness ofNonprofit Boards.” Nonprofit and Voluntary Sector Quarterly, 1998,27(2), 159–182.

Johnson, J. L., Daily, C. M., and Ellstrand, A. E. “Boards of Directors:A Review and Research Agenda.” Journal of Management, 1996, 22,409–438.

Kiel, G. C., and Nicholson, G. Boards That Work: A New Guide forDirectors. Sydney, Australia: McGraw-Hill, 2003.

Leblanc, R. “Assessing Board Leadership.” Corporate Governance,2005, 13(5), 654–666.



McGregor-Lowndes, M. “Nonprofit Corporations: Reflections onAustralia’s Largest Nonprofit Insolvency.” Australian Journal of Cor-porate Law, 1995, 5(4), 417–441.

Morrow, L., and Bartlett, L. Contrary and Congruent Views of Leader-ship and Management in the Australian Social Economy. Sydney,Australia: Social Economy Executive Education Network, 2007.

Nicholson, G. J., and Kiel, G. C. “A Framework for DiagnosingBoard Effectiveness.” Corporate Governance, 2004, 12(4), 442.

Nicholson, G., and Newton, C. J., “The Role of the Board of Direc-tors: An Exploratory Study of the Perceptions of ManagerialElites.” Journal of Management and Organization, 2010, 16(2),204–218.

Nicholson, G., Newton, C., and Sheldrake, M. L. “Key GovernanceIssues for Funded Nonprofit Organisations in Queensland.” Work-ing Paper No. CPNS 41. Brisbane: Australian Centre for Philan-thropy and Nonprofit Studies, 2008.

Nobbie, P. D., and Brudney, J. L. “Testing the Implementation, BoardPerformance, and Organizational Effectiveness of the Policy Gov-ernance in Nonprofit Boards of Directors.” Nonprofit and VoluntarySector Quarterly, 2003, 32(4), 571–595.

Pfeffer, J. “Size, Composition and Function of Hospital Boards ofDirectors: A Study of Organization Environment Linkage.” Admin-istrative Science Quarterly, 1973, 18, 349–364.

Pye, A. “Studying Board Context, Process and Dynamics: Some Chal-lenges for the Future.” British Journal of Management, 2005,16(Suppl. 1), S27–S38.

Salamon, L. M., and Chinnock K. “So What? Accounting for Non-profit Impact.” Accountability Forum, 2004, 2, 26–30.

Slesinger, L. H., Self-Assessment for Nonprofit Governing Boards.Washington D.C.: National Center for Nonprofit Boards, 1991.

Smith, P. C., and Richmond, K. A. “Call for Greater Accountabilitywithin the US Nonprofit Sector.” Academy of Accounting and Finan-cial Studies Journal, 2007, 11(2), 75–88.

Sonnenfeld, J. A. “What Makes Great Boards Great?” Harvard Busi-ness Review, 2002, 80(9), 106–113.

Stiles, P., and Taylor, B. Boards at Work: How Directors View TheirRoles and Responsibilities. New York: Oxford University Press,2001.

van Ees, H., Gabrielsson, J., and Huse, M. “Toward a Behavioral The-ory of Boards and Corporate Governance.” Corporate Governance,2009, 17(3), 307–319.

Wageman, R., Hackman, J. R., and Lehmann, E. “Team DiagnosticSurvey: Development of an Instrument.” Journal of Applied Behav-ioral Science, 2005, 41(4), 373–398.



GAVIN NICHOLSON is an associate professor at the Australian Centre ofPhilanthropy and Nonprofit Studies, School of Accountancy, QueenslandUniversity of Technology, Australia.

CAMERON NEWTON is an associate professor at the School of Manage-ment, Queensland University of Technology, Australia.

MYLES MCGREGOR-LOWNDES is a director of the Australian Centre of Phil-anthropy and Nonprofit Studies, School of Accountancy, QueenslandUniversity of Technology, Australia.





Appendix 1: Survey Items from Our Study

Construct Item*

Defined (i.e., clearly Everyone knows who is a member of our board, and who is not.defined board) People who know this board could name all its members.

It would be nearly impossible to accurately name who is and who is not a memberof the board in this organization.**

Interdependent (i.e., board There is little need for board members to work together on this board.**members rely on each other Our board’s success relies on much coordination and communication between board and work together) members.

On this board, the nature of our roles or tasks requires board members to rely heavily on each other.

Clear (i.e., the objectives or Every member of this board knows the board’s purposes and roles—what it is goals of the board are clear) here to accomplish.

It would be difficult to outline precisely what the purposes or roles of the board is in this organization.**There is a clear delineation in this organization between the roles of the boardand the roles of management.

Consequential (i.e., the board’s This board’s roles are of great consequence for the organization.role makes a difference) This board’s roles don’t make much difference to the organization.**

Diversity (i.e., composition This board has the right mix of members with a diverse range of skills and experi-has the right diversity) ences required of the group.

The membership of this board is too diverse—people are so different that they don’t work well together.**There isn’t a sufficiently wide range of perspectives and experiences on this board ifwe are to carry out our roles.**

Skills (i.e., the knowledge, Between them, board members have the necessary knowledge, experience, and skills, and experience of the skills to carry out the board’s roles and achieve its goals.board are appropriate) Some board members do not possess the knowledge, skills, or experience required to

contribute to the board’s work.**All board members have knowledge, skills, and experience that contribute to theboard’s work.

Information (i.e., the board Board members find it easy to get the information they need to carry out their roles has access to the information or tasks.it needs) The board has difficulty accessing the information we need to carry out our

roles or tasks.**The board often finds itself unaware of information it needs to carry out its role.**The board can get the information it needs to carry out its role.

Resources (i.e., the board has On the whole, the board is provided with appropriate resources for the job the resources it needs) required of it.

There is a definite lack of resources for the board considering the role requiredof it.**

Task (i.e., board members Board members help motivate each other and stimulate greater commitment to the provide each other with board and organization.feedback on their task Board members act to ensure the board continually develops and takes the most performance) effective approach to its role.

Board members act to ensure all members’ skills, experience, and knowledge are used.



Unhelpful (i.e., board members Board members tell other board members what they should do.provide unhelpful feedback on Board members tell other board members how they should do tasks.performance)

Psychological Safety (i.e., board If you make a mistake on this board, it is often held against you.**members can speak their mind) Members of this board are able to bring up problems and tough issues.

People on this board sometimes reject others for being different.**It is safe to take a risk on this board.It is difficult to ask other members of this board for help.**No one on this board would deliberately act in a way that undermines myefforts.When working with members of this board, my unique skills and talents arevalued and utilized.There are certain issues/matters which are off-limits for discussion.**

*Italicized items indicate a derivation of the TDS (Wageman, Hackman, and Lehmann, 2005).

**Indicates a reversed item.

the nonprofit board as a team - adea.com.au · general corporate governance literature recognizes...

Documents